Adolescent alcohol abuse is associated with continued substance abuse into adulthood, and the public health burden of adolescent drinking is evident in its association with drunk driving, unwanted and risky sexual behavior, truncated educational achievement, altered life trajectories, and premature death. To reduce that public health burden with appropriate interventions, we must understand why these adult outcomes associate with adolescent alcohol exposure. The obstacles in drawing causal inferences from observational data are well-known; potential confounds in the association of adolescent drinking with adverse adult outcomes are many, and the available data do not permit confident causal inferences. Using the unusual resources available to us from decades of longitudinal research with Finnish twins, we propose studies designed to document the consequences of adolescent alcohol use across the lifespan, comparing outcomes in exposure-discordant twin pairs to address the question of whether adult consequences of adolescent drinking are causal in nature, or whether, instead, they arise from selection processes and correlated liabilities. Exposure-discordant co-twins, critically including adequate numbers of exposure-discordant monozygotic co-twins, can yield incisive information with which to evaluate this question. We will obtain and analyze data from four large Finnish cohorts longitudinally studied for substance use from entry ages 12-22, tracking adult outcomes to ages 25-59. Our research plan will yield informative data for individual-based analyses on 22,000 individuals and robust analyses of exposure-discordant co-twins on 1,250 twin pairs, including a substantial subset of the most informative: monozygotic co-twins who are discordant for their alcohol exposure in adolescence. Results of this innovative research will provide important information on the meaning of long-term adult associations with adolescent alcohol abuse, inform intervention efforts, and address a major public health burden.